[02:08:39] ikogan joins the room [06:53:24] Jeff Ward joins the room [07:29:49] Joseph Ishac joins the room [07:50:59] alexmcmahon joins the room [07:55:54] ikogan leaves the room [08:02:19] ikogan joins the room [08:17:34] Will Ivancic joins the room [08:26:58] Armando Caro joins the room [08:43:59] teemu joins the room [08:49:40] sftcd joins the room [08:50:44] Hi all! Alex and I are in the terminal room, other folks are downstairs in the little theater until noon [08:54:40] yep where here in the rather large 'little theater' [09:10:12] If Elwyn's there, his coffee's up here [09:30:58] Hi Joe /will , not having much luck getting to dtn://unit017.dtn/ping from dtn://basil.dsg.cs.tcd.ie.dtn . Would appreciate if someone could add routes / links from unit017 to dtn://basil.dsg.cs.tcd.ie.dtn / basil.dsg.cs.tcd.ie:4556 and routes to dtn://dtngateway-2-200.dtnbone.ocp.ohiou.edu.dtn, dtn://dtnmule-2-10.dtnbone.ocp.ohiou.edu.dtn, dtn://dtnbone-2-31.dtnbone.ocp.ohiou.edu.dtn & dtn://sphere.dtnbone.ocp.ohiou.edu.dtn via dtn://basil.dsg.cs.tcd.ie.dtn [09:32:44] * once configure you should be able to reach all nodes except dtn://dtnmule-2-31.dtnbone.ocp.ohiou.edu.dtn ( we are working on this) [09:39:29] I ran ldconfig and I see ld.so.conf in /etc But I still get cannot find ldtnapi [09:39:52] Alex - here is the link to OU's compatability applicaitons [09:39:54] will: https://ion.ocp.ohiou.edu/SBP-API.tar.bz2 with documentation at https://ion.ocp.ohiou.edu/sbpdoc/index.html [09:42:36] alexmcmahon: I'm online here... I'll work on adding that info [09:52:11] alexmcmahon: added the routes [09:52:19] can't seem to ping [09:53:19] I can't even seem to ping ping basil.dsg.cs.tcd.ie [09:53:34] is it alive? I have all the hosts going through that link [09:54:12] (that's normal pings that aren't working, not dtnpings) [10:19:38] any tests here, I'd like to take both ION and DTN2 down on unit017 for a sec? [10:27:09] some icmp is blocked on basil.dsg.cs.tcd.ie but you should be able to connect - you should be able to telnet to port 4556 [10:28:10] Hans, has anyone been using the SBP Applications? If so, has it proven useful? [10:31:45] no yet.. [10:39:00] teemu leaves the room: Computer went to sleep [10:39:31] Jeff Ward leaves the room [10:41:01] We are losing our room and it is past lunch. Many will move to the terminal room or elsewhere in the Convention Center. [10:41:09] Armando Caro leaves the room [10:41:46] Will Ivancic leaves the room: Computer went to sleep [10:42:37] Joseph Ishac leaves the room [11:09:54] Will Ivancic joins the room [11:23:51] @alex: what what the wrap's IP address again? [11:25:29] guess alex isn't looking at jabber - can someone else resolve dtnmule-2-31.dtnbone.ocp.ohiou.edu? [11:25:58] (reason is I'm in a session and haven't got my cheat sheet of addresses;-) [11:42:38] hkruse joins the room [11:47:19] hi hans [11:48:00] Hello -- good to hear from you [11:48:31] dtnmule-2-31.dtnbone.ocp.ohiou.edu has address 130.129.53.239 [11:48:47] from http://lin5.its.ohiou.edu:8080/name-list [11:49:05] but may have new address today. [11:50:35] I can ping dtnmule-2.31 [11:56:22] thanks will [11:56:36] beroset joins the room [12:00:00] That is IP ping. I haven't tried dtnping [12:00:28] no need; still building dtn2 for openwrt so no daemon running there [12:01:49] I could not compile SBP: the ldtn_api error occurs when either 1) the .so isn't in a directory the machine knows to look (check ld.so.conf) or 2) ldconfig hasn't been run [12:02:28] The system could not find ldtnapi [12:03:11] GRC unit017 was running and older version of DTN2 - apparently. Could this be the issue? [12:06:02] anyrhine joins the room [12:06:31] Armando Caro joins the room [12:07:11] Joseph Ishac joins the room [12:15:54] Jeff Ward joins the room [12:16:28] teemu joins the room [12:35:17] hkruse leaves the room [12:52:30] Jeff Ward leaves the room [12:52:30] Armando Caro leaves the room [13:02:35] Will Ivancic leaves the room [13:04:26] ostermann joins the room [13:04:47] hkruse joins the room [13:05:18] beroset leaves the room [13:06:02] Armando Caro joins the room [13:06:02] Jeff Ward joins the room [13:06:56] Joseph/anyone else: We just did a 4 node inline test with 3 implementations, and we'd like to add more. anyone want to join in? [13:07:19] sure [13:07:29] I just upgraded u17 [13:07:29] We did pings from TKK Java -> BBN -> BBN -> TKK iPhone [13:07:48] and pinged bbn-*-m10 [13:07:53] that's a dtnping [13:07:54] is that the one that had dtn2 old version? [13:08:02] it had 2.6.0 [13:08:12] but I pulled the cvs [13:08:37] should we insert u17 in between our two bbn nodes? [13:08:49] we're going to try pinging u17 [13:09:13] sounds fine [13:10:39] You could extend the chain to dtn;//quest.dtn/echo (ION TCP) or dtn://testbed4.dtn/echo [dtn://testbed4.dtn/echo] (ION UDP) (as long as we have a static route back). Or send one-way to the /sink EID and check the result in our log display (http://quest.its.ohiou.edu [http://quest.its.ohiou.edu]) [13:11:25] I'd much rather test a disconnection scenario [13:11:50] has anyone tried that? I almost had it [13:13:01] for disconnected scenarios we have a few test networks with simulated disconnections running at the lab which could be used [13:13:06] what are you looking for [13:13:26] Joseph Ishac: which port should I be sending to on unit017 now? [13:14:12] well earlier when I saw teemu shut down his laptop, I sent a request to his camera from my laptop. the request went to unit017 and was held there. I then shut down my laptop [13:14:37] just now when I came back online I had noted that I had a picture waiting for me at unit017 [13:14:51] however my local dtn node crashed everytime it tried to grab it [13:15:25] Jeff Ward: For dtn2 the ports are 5017... for ION 4556 (TCP and UDP on both) [13:15:58] bpecho should run on /echo; and dtnping for /ping [13:16:43] OU: I am restarting quest.dtn to install new static routes [13:16:49] So I claim I had a partial success with a fully disconnected scenario, where there was never a fully connected path. [13:17:04] Joseph Ishac: ION's TCP CLA is not compatible with DTN2's, correct? [13:17:30] They are compatible as long as you have the lastest versions of both [13:17:37] True, I don't know which version of TCPCLI is running on ION, I believe it to be the lightweight one [13:18:01] Jeff Ward: hkruse: it should be the latest ion... lemme check [13:18:46] it's running ion-2.0_r245_ou710 [13:19:24] That is what we are running - tcpcl is good with DTN2 [13:19:31] stcpcl is the light-weight one [13:19:52] DTN2 needs to be from the mercurial repo (needs a tcpcl patch) [13:20:42] a induct tcp 192.55.90.165:4556 tcpcli ; is in the config [13:21:08] but that doesn't mean dtnping will work [13:21:23] you need the api tools for that IIRC [13:21:31] No. We are running echo out of the SBP [13:21:54] the ION stock bpecho confuses dtnping, I think [13:22:15] so Jeff you should be able to dtnping unit017 at port 5017... I was able to hit your box [13:22:46] not at the moment it seems though [13:25:00] Hmm, I still have an issue with dtn2 on my laptop... I'll start to get "fatal error opening dtn handle: error in ipc communication" at the console and [13:25:32] ASSERTION FAILED (!incoming_.empty()) at conv_layers/StreamConvergenceLayer.cc:1184 STACK TRACE: 0x8329017 0x80e1a33 0x80e525a 0x80ea994 0x81702f0 0x8354bc7 0x8354d69 0xb7fc44ff [13:25:34] in the log [13:25:50] I need to track this anoying bug down [13:25:59] That's why we made our own BPA. [13:26:27] Understood :) [13:27:44] Jeff Ward: I am pinging your box again.... $ dtnping -c 5 -e 10 dtn://bbn-dtnbone-m10/ping source_eid [dtn://unit017.dtn/ping.3450] dtn_register succeeded, regid 12 PING [dtn://bbn-dtnbone-m10/ping] (expiration 10)... 20 bytes from [dtn://bbn-dtnbone-m10/admin]: 'dtnping!' seqno=0, time=236 ms bundle deleted at [dtn://bbn-dtnbone-m10/admin] (no additional information): seqno=0, time=375 ms 20 bytes from [dtn://bbn-dtnbone-m10/admin]: 'dtnping!' seqno=1, time=209 ms bundle deleted at [dtn://bbn-dtnbone-m10/admin] (no additional information): seqno=1, time=367 ms 20 bytes from [dtn://bbn-dtnbone-m10/admin]: 'dtnping!' seqno=2, time=193 ms bundle deleted at [dtn://bbn-dtnbone-m10/admin] (no additional information): seqno=2, time=356 ms 20 bytes from [dtn://bbn-dtnbone-m10/admin]: 'dtnping!' seqno=3, time=324 ms bundle deleted at [dtn://bbn-dtnbone-m10/admin] (no additional information): seqno=3, time=332 ms 20 bytes from [dtn://bbn-dtnbone-m10/admin]: 'dtnping!' seqno=4, time=194 ms bundle deleted at [dtn://bbn-dtnbone-m10/admin] (no additional information): seqno=4, time=352 ms [13:28:40] wang.russell77 joins the room [13:28:43] That looks normal [13:29:27] Jeff Ward: yep... hopefully you can ping it though.... [13:30:02] I can ping unit017 through DTN2 at port 5017 [13:30:57] However, my pings over TCP to port 4556 go unanswered. Will ION reply to bundles going to /ping? [13:31:40] Will Ivancic joins the room [13:32:26] No... you can send message to /sink I believe [13:32:56] Will that echo, or is it really a sink? [13:33:47] That is probably a sink to the console [13:34:40] Joe should be able to start bpecho, but that returns a fixed reply (rather than echoing the bundle back). [13:34:46] Jeff Ward: what hkruse said... it sinks to something we can read at the console. We don't have OU's tools built - Will wasn't able to succesfully compile them yet [13:35:18] bpecho is running, but on /echo not /ping [13:35:18] dyoung-ohiou joins the room [13:41:24] Can ION on unit017 (or some other node) work as a forwarder sitting between two other nodes? [13:41:58] yes [13:42:43] so long as I load ION with the necessary route info [13:42:46] OU: still working on quest.dtn -- coordinate any work with me [13:43:04] Can we put it between bbn-dtnbrick-05 and bbn-dtnbrick-10? [13:44:54] anyrhine leaves the room [13:45:15] hi Will: any luck getting those SBP tools to compile? [13:46:24] hkruse: found a config file problem with mcmahon's network on quest; should be fixed now. [13:47:07] We got in each others way -- just had to stop ION [13:47:42] hmm? I only edited the log file; didn't restart the node or anything. [13:47:56] *log = config file... [13:48:14] OK, I will restart [13:48:44] ll [13:50:34] I think jeff meant: bbn-dtnbone-m05 and bbn-dtnbone-m10 [13:51:16] wang.russell77 leaves the room [13:54:07] no luck getting SBP tools to compile. Want's file ldtnapi but I cannot find it or make it. [13:54:28] Sorry, I guess I am stuck in the last phase of our project. Wrong node names. [13:55:50] gregc joins the room [13:57:55] Will: ldtnapi should refer to libdtnapi.so; it should build when DTN2 builds, and get installed when 'make install' is run. anyway, if you look in /applib, there should be a libdtnapi-2.6.0.so or something like that; that's what -ldtnapi is looking for. [14:00:37] so I'm trying to run OU's script [14:01:07] for the API tools.... why is is asking me for root when I built and all specified locations are in my home directory :) [14:01:43] it's asking for root to install the tools; I'll find the specific lines [14:01:55] Joseph, where R you? I am next to the terminal room [14:02:15] I'm in the plenary room [14:03:04] Besides the camera application, what other applications have been run. I want to put that in tomorrow's report. [14:03:24] That is why I want to know if the OU API is working for people [14:03:41] Joseph: it's an oversight on my part; I should try to copy the files without sudo first, then ask for root [14:04:10] We are running the SBP versions of echo and sink on all OU machines [14:04:45] gclark-ohiou: yeah that should work [14:05:16] But has anyone been else been using them? BBN or TTK or Trinity? [14:07:12] Joseph: if you don't provide the root password, the installer will warn that it couldn't copy the files and keep going. . . you'll have to copy the .so files in the bin directory and sbp_api.h and sbp_socket.h from the inc directory to somewhere the tools will be able to find them when they build [14:07:25] Hans, has anyone been communicating over the UDP disconnected restricted links? [14:08:19] Will: to my knowledge, no one else has used these tools, no. The release for the disconnectathon wasn't meant to be a stable version, just a release of what we were running on our nodes. [14:08:30] gclark-ohiou: will a simple make after the copy work then? [14:09:03] Joseph: the make should've already been completed; look in sbp_api/bin [14:09:30] gclark-ohiou: yeah but after the failed copies it tried to make the tools (or something) [14:10:00] Will: hard to tell; I am not aware of someone outside OU doing so. [14:10:17] Joseph: ahh, yes. it goes into each tool's subdirectory and either runs 'make ion' or 'make dtn' [14:10:45] I'd say the vast majority of testing (as per the plans submitted) involved implementation interoperability, and routing [14:11:47] static routing. [14:12:10] And flooding (Teeemu) [14:12:20] yes, i have that. [14:12:39] any disconnection tested yet? [14:12:46] someone want to bring me a cookie? [14:12:46] Joseph: depending on which version of the library was compiled [14:12:48] :) [14:13:30] I know Joseph was attempting to do this with Teemu [14:13:41] From OU: we have focused on network monitoring and diagnostics -- lots of data and insight (non processed yet) [14:14:33] I saw his account earlier about the image retrieval across shutdowns of two of the machines involved [14:14:33] I have that too. I will post my draft slide to the discconnectation N4C site during the plenary. [14:15:03] Joseph, what kind of cookie and where are you sitting. I'll come down. I skipped lunch and need a cookie [14:15:57] Do you want a drink too? [14:16:03] Will Ivancic: Anything with sugar works or grab something and find me and I'll grab my own :) [14:16:30] It looks like the snack options are chocolate cookie, chocolate brownie, or chocolate cupcake (I think it is all made from the same substance) [14:17:03] Will Ivancic leaves the room: Computer went to sleep [14:23:43] Joseph: having any luck? [14:23:54] Will Ivancic joins the room [14:24:06] I'm off to the plenary so my node will be down [14:24:47] teemu: have you patched your camera dtn2 node? [14:25:17] no, it's still the old version [14:25:42] OK [14:26:09] teemu leaves the room: Computer went to sleep [14:30:07] BOXING-PC joins the room [14:30:08] gclark-ohiou: almost there... everythings coppied, although ld isn't finding the libraries when I go to make in the tools directory. I have /home/usr/lib in LD_LIBRARY [14:30:26] Joseph: try ldconfig [14:34:21] tried ldconfig /home/usr/lib [14:34:51] didn't seem to help, maybe I'm not as familiar with ldconfig as I thought... time to hit the man page [14:35:44] Joseph: one other thing to check. . . did you build SBP for ION, DTN2, or both? [14:36:23] both [14:36:37] Joseph: all righty. so much for that idea then :) [14:36:51] teemu joins the room [14:37:35] my dtn camera is now sending pics from the plenary [14:39:22] teemu: is your node also on udp, or tcp only? [14:39:31] udp and tcp [14:39:33] same port [14:40:28] great [14:41:20] David: if you can get this to work from testbed4, try from 3 or 5 as well [14:42:16] will do- i'll be commenting out the tcp configurations on quest and instead forward to testbed4 for the camera. [14:44:03] teemu: can you modify your routes to go to OU machines via testbed4.dtnbone.ocp.ohiou.edu udp:4556 and not quest? [14:44:07] ok so I did it the old fashion way [14:44:24] by modifying the make file :) [14:44:35] I can open a link to testbed4.dtnbone.ocp.ohiou.edu, routing is flooding [14:45:30] OK, also remove the tcp link to quest [14:45:30] link is added now [14:45:53] thanks, testbed4 is actually dtn://testbed4.dtn [14:46:04] Joseph: heh, that works too :) [14:46:20] OU folks: I'm editing the config files on testbed4 and quest [14:46:29] Sorry for spamming the testbed4 log file, i'm messing around with bptrace. Anyone know which endpoint trace should send to? [14:48:17] dtn://testbed4.dtn/trace but it doesn't look like anything is listening on that eid right now [14:49:02] Greg: you can send to any endpoint I think; try ack or echo [14:49:11] consult: http://lin5.its.ohiou.edu:8080/testbed4-status to see routes, endpoints etc- and the endpoints have a pid of the process listening on it. [14:54:11] debugging code blows over a saturated IETF network [14:55:00] OU: editing done on testbed4 and quest; someone restart ION on them (and msg in here before you do it/make sure nobody else already did) [14:55:31] Unit017 is now crashing too... anyone want to help debug the error I put up earlier? [14:56:08] Joseph: which error? [14:56:12] it seems to be a fairly fatal flaw [14:56:31] repost please; i've lost my backlog. [14:56:37] ASSERTION FAILED (!incoming_.empty()) at conv_layers/StreamConvergenceLayer.cc:1184 STACK TRACE: 0x8329017 0x80e1a33 0x80e525a 0x80ea994 0x81702f0 0x8354bc7 0x8354d69 0xb7fc44ff [14:57:06] most of my debug environment is on my remote machine [14:57:06] ... oh no. [14:57:15] Joseph: check in sbp_echo.c [14:57:37] gclark-ohiou: this was before the sbp tools [14:58:12] joseph: okay. *wipes forehead* [14:58:26] it seems to like crashing when other agents try sending it bundles [14:59:01] this is DTN2 [14:59:44] Joseph: right. I was having this issue when I was sending a bundle back that was bigger than what the receive apps in DTN2 were expecting [14:59:59] Joseph: which is why I immediately assumed sbp_echo [15:00:26] OU: quest and testbed4 will reload ION shortly [15:02:32] OU: quest and testbed4 are back [15:03:23] So an older version of DTN2 didn't appear to have this issue... (but had other issues that prevented us from talking to spindle) [15:05:09] Greg: any luck with bptrace? [15:06:47] This network is not neutral... it's clearly scheming against us... ;) [15:07:28] Joseph Ishac wonders how many times my net connection will drop during this plenary [15:08:23] Joseph: heh, that bad, huh? [15:08:59] gclark-ohiou: yes [15:19:30] ikogan leaves the room [15:19:30] teemu leaves the room [15:19:31] gclark-ohiou leaves the room [15:19:31] gregc leaves the room [15:19:32] alexmcmahon leaves the room [15:19:32] josh leaves the room [15:29:30] David: did you get to Teemu's machine? [15:30:20] Joseph Ishac leaves the room [15:31:08] Not yet. [15:32:48] teemu, would you attempt to send a bundle to dtn://quest.dtn/sink and dtn://testbed4.dtn/sink ? [15:33:12] and... i didn't read the log that says he's not here. [15:37:08] Joseph Ishac joins the room [15:37:08] BOXING-PC leaves the room [15:37:08] dyoung-ohiou leaves the room [15:37:08] hkruse leaves the room [15:37:08] ostermann leaves the room [15:37:08] Will Ivancic leaves the room [15:37:08] sftcd leaves the room [15:42:02] sftcd joins the room [15:42:31] Will Ivancic joins the room [15:44:11] teemu joins the room [15:45:47] josh joins the room [15:46:01] ikogan joins the room [15:46:12] dyoung-ohiou joins the room [15:46:45] teemu: could you send a text message to dtn://tetbed4.dtn/sink ? [15:47:53] typos are getting me dtn://testbed4.dtn/sink [15:48:05] just a sec [15:49:21] ikogan leaves the room [15:50:03] ok sent [15:50:35] Jeff Ward: your wireless cutting in and out? Mine is pathetic - dozen or so times already? [15:51:02] Mine has been solid sitting in the terminal room all afternoon. [15:51:20] Sitting in the plenary ... not so lucky [15:51:52] wireless in plenary has been fine for me, except Jabber keeps dying [15:51:59] missing bad-attitude :/ [15:52:32] gclark-ohiou joins the room [15:52:44] My whole wireless drops.... I bame the wireless drivers in linux [15:52:57] making managing these dtn nodes imposible [15:53:15] teemu, what endpoint is your node listening to for pings? [15:53:32] also, your test was successful. [15:53:48] dtn://teemu.comnet.tkk.fi/ping [15:55:32] ok, I can ping you and that functions... so why isn't the camera working... [15:55:52] ugh- udp convergence layer- this won't work. [15:56:17] maybe the pic won't fit into a single udp packet [15:57:20] yeah. do your logs indicate a failure in udp from the camera? [15:58:04] I'm requesting from dtn://testbed4.dtn/sendfile [15:59:02] yep, that's the problem [15:59:05] [1248969478.495726 /dtn/cl/udp/sender/0x3f68000 error] send_bundle: error sending bundle (wrote -1/29298): Message too long [16:00:01] Roy joins the room [16:07:55] ikogan joins the room [16:24:30] ikogan leaves the room: Replaced by new connection [16:24:31] ikogan joins the room [16:54:52] Joseph Ishac leaves the room [17:12:37] Roy leaves the room: offline [17:18:36] teemu leaves the room [17:20:12] sftcd leaves the room [17:20:50] Will Ivancic leaves the room: Computer went to sleep [17:21:47] ikogan leaves the room [17:23:54] teemu joins the room [17:38:18] ikogan joins the room [17:45:58] Jeff Ward leaves the room [17:48:51] Armando Caro leaves the room [17:58:38] hkruse joins the room [17:58:42] hkruse leaves the room [18:10:11] teemu leaves the room: Computer went to sleep